Embed Size (px)
Citation preview
FULL RESEARCH ARTICLE
Strength analysis of aluminium foil parts made by compositemetal foil manufacturing
Javaid Butt1 • Habtom Mebrahtu1 • Hassan Shirvani1
Received: 24 December 2015 / Accepted: 21 March 2016 / Published online: 5 April 2016
� Springer International Publishing Switzerland 2016
Abstract A number of different parts for mechanical
testing were produced using composite metal foil manu-
facturing (CMFM). This process is a combination of lam-
inated object manufacturing and brazing technology.
Aluminium is one of the toughest metals to join and
CMFM can achieve this task with ease. By using alu-
minium 1050 foils of 0.1 mm thickness, various parts were
made according to British and International Standards,
including lap joints, peel specimens, dog-bone specimens
and tested for their mechanical properties. A special, 80 %
zinc and 20 % aluminium by weight, brazing paste was
utilized for joining the foils together. The test of the single
lap joints show that none of the specimens failed at the
bonded area and the failure was always due to fracture of
the parent metal. Cohesive failure was also observed for
the single lap joints by using 10 mm thick aluminium metal
plates. It helped in calculating the lap shear strength which
is a useful design parameter. The peel test showed good
bond consistency in all the specimens with an average peel
strength of 20 MPa. Comparative tensile test was con-
ducted with a dog-bone specimen machined from a solid
block of aluminium 1050 and specimens made with
CMFM. The results showed that the specimens made by
CMFM fracture at force values that are higher than that of
the parent metal. This demonstrates that CMFM has the
capability to produce high quality and stronger parts as
compared to conventional machining methods employed
for the production of metal parts. The effect of using dif-
ferent number of layers for the same cross-sectional area
has also been investigated.
Keywords Additive manufacturing � Brazing �Laminated object manufacturing � Lap-shear testing �Metal foil � Metal parts � Peel test � Tensile testing
1 Introduction
Additive manufacturing (AM) of metal parts provides
solutions to real-world problems as they can be tested
under similar conditions to show their behaviour. A large
number of AM processes are commercially available that
use plastics, polymers, photopolymers, ceramics, etc. to
produce parts [1–5]. These parts mainly serve as prototypes
and do not give any insight into the design parameters.
There has always been an emphasis on the production of
metal parts using AM because it provides an environment
for direct testing. Furthermore, AM methods can produce
complex geometries like conformal cooling channels for
injection moulding, which are not possible with conven-
tional machining/subtractive techniques. A couple of
commercially available AM processes capable of produc-
ing metal parts are direct metal laser sintering (DMLS) and
electron beam melting (EBM). They use powder metals
and have been widely researched over the years [6–10].
Although these technologies are capable of producing good
quality parts, they still do not ease the pressure when it
comes to the production of metal parts because of the very
high cost of their machines and inability to cope with the
material requirements [11, 12]. They have a number of
& Javaid Butt
Habtom Mebrahtu
Hassan Shirvani
1 Engineering & Built Environment, Anglia Ruskin University,
Bishop Hall Lane, Chelmsford CM1 1SQ, UK
123
Prog Addit Manuf (2016) 1:93–103
DOI 10.1007/s40964-016-0008-5
limitations like materials, particular grain size of powder
metal to achieve certain degree of mechanical integrity,
size of parts being made, build speed, surface finish, etc.
Also, after production, the powder removal process
requires more time and thus reduces the efficiency of the
process [13, 14]. Therefore, efforts were made to overcome
these limitations and other AM technologies were pursued.
This idea led to the modification of laminated object
manufacturing (LOM). The LOM method originally made
use of paper sheets with adhesives on one side but since the
setup was cheap, further materials including plastic and
metal laminates were experimented with, to enhance the
capabilities of this technology [15]. One such method is
known as metal foil LOM (laminated object manufactur-
ing) and uses metal sheets for the production of parts. It
was an experimental approach that was aimed at the pro-
duction of large moulds. For the structural stability of the
parts, Metal Foil LOM can only work with foils having a
thickness of more than 0.5 mm as anything less than that
results in a significant staircase effect. The produced parts
are not of very good quality either and have a poor surface
finish. Post processing is required to improve the
mechanical properties of the parts which makes the process
inefficient [16, 17]. Epoxy adhesives and glues have also
been used to join metals sheets together using LOM but the
produced parts are far from perfect in terms of their
material properties and in some cases serious failures have
occurred [18, 19]. Another method that has shown promise
is known as ultrasonic consolidation (UC). It has shown
great promise in working with difficult metals like alu-
minium and stainless steel. It combines ultrasonic seam
welding of metals and layered manufacturing techniques to
build up a solid freeform object. The process uses a
sonotrode connected to a transducer to apply pressure and
ultrasonic oscillations for bonding. The process is repeated
until the required height is achieved and then a CNC
(computer numerical control) mill is used to trim the excess
foil from the component and achieve the required geome-
try. Another finishing mill is brought into action to create
the required tolerance and surface finish. After the trim-
ming and finishing, the finished part is removed from the
anvil [20, 21]. The biggest challenge of this process is
optimization of the process for bond density and plastic
flow to have a better contact between the foils [22]. All of
these processes have been researched upon and parts have
been produced using them but they are limited in what they
can do and achieve in terms of producing metal parts.
In addition to making metal parts by alternative and
cheaper methods, the production of composite materials
has been receiving unprecedented attention from the aca-
demia, mainstream media, investment community and
national governments around the world. Metal composites
or metal matrix composites (MMCs) are gaining popularity
because a number of industries rely on them to provide
cheaper, lighter, and stronger alternatives. AM processes
are not far behind in this endeavour either and a number of
them have produced MMCs including selective laser
melting/sintering (SLM/SLS), laser engineered net shaping
(LENS), three-dimensional printing (3DP), LOM and UC.
However, it required a great degree of additional equip-
ment and time to produce a metal composite using AM
processes and their mechanical integrity in some cases is
far from ideal [23–25].
This study clearly indicates that there is a gap that needs
to be filled when it comes to the production of cheap and
high-quality metal parts. The limitations related to con-
ventional metal AM methods have to be minimised to
make way for better and efficient product development.
Another process has joined the list of metal AM methods
capable of using metal foils for the production of functional
parts and is termed as composite metal foil manufacturing
(CMFM). It combines the simplicity of LOM with the
flexibility of brazing. This integration makes the process
adaptable to changes. A model of a machine based on the
principle of CMFM is shown in Fig. 1. The product
development has two stages—the principles of LOM for
cutting and placing of sheets and brazing to join the foils
and produce a part. Brazing is similar to soldering, but the
main difference lies in the operating temperatures as
joining operations that are under 450 �C are considered to
be soldering and those above 450 �C lie under brazing.
Copper, silver, and gold are easy to braze because they
have good braze-ability. Iron, mild steel and nickel are next
in difficulty. Because of their thin, strong oxide films,
stainless steel and aluminium are even more difficult to
braze. A detailed description of the process has been pre-
sented in our previous research along with the testing of
copper foils to establish the effectiveness of the process
Fig. 1 Composite metal foil manufacturing process
94 Prog Addit Manuf (2016) 1:93–103
123
[26, 27]. To demonstrate the flexibility of the proposed
process, 0.1 mm thick Aluminium 1050 grade foils with
H14 � hard temper were chosen for the research work.
Aluminium is one of the hardest metals to braze and var-
ious aluminium alloys have different braze-ability: 1xxx,
2xxx, 3xxx, 4xxx, and 7xxx are easier to braze than the
6xxx series alloys. Magnesium content in the 5xxx series
alloys makes them the most difficult to braze. It is because
in addition to the oxide layer of aluminium, the oxide layer
of magnesium needs to be removed as well to perform the
brazing process. The magnesium oxide layer also forms on
contact with air so instead of one, now there are two layers
to remove for joining aluminium alloys containing mag-
nesium [28]. The removal of the oxide layer is the key to
the brazing process as once it is removed; the process can
go on easily. Aluminium oxide is not so easily removed
and requires stronger fluxes that can go up to temperatures
of 550 �C. In case of very thin aluminium foil, which is the
case for this research work (0.1 mm), there is a danger of
pitting so proper care must be given in dealing with such
cases. For these reasons, 80 % zinc and 20 % aluminium
by weight brazing paste with non-corrosive flux was cho-
sen to produce aluminium specimens. The melting point of
the brazing paste in use ranges between 410 and 470 �Cand it becomes liquid in this range. However, it should not
be kept at these temperatures for longer periods of time as
the flux would burn off and the paste would not be able to
penetrate the tenacious oxide layer on the surface of the
aluminium foil.
2 Experimental procedure
CMFM is an extremely complex process and a number of
different components have to work together in complete
harmony to produce parts. That is why the complex process
was broken down into simple independent steps for prac-
ticality. An experimental procedure was devised to create
metal parts using metal foils. The use of metal foils makes
the production of parts very cheap as compared to other
AM methods. The process of brazing can be tricky for
aluminium; therefore, it was essential that proper care was
given to every specimen. The ability of the process to work
with difficult metals such as aluminium is what makes it
flexible as well as adaptable; two characteristics that are
very important in a metal parts manufacturing system.
The foils of aluminium were used as supplied without
any surface treatment as it takes time and one of the key
features of the process is that it reduces the production time
to a great extent compared to other AM methods. The foils
were cut and then deposited with brazing paste. Every
specimen was sandwiched between two stainless steel
plates fitted with nuts and bolts. At any given time, only
one specimen was placed inside the plates and a uniform
layer of paste was achieved by tightening the nuts with a
torque wrench. The use of torque wrench ensured that
repeatability can be observed in all the specimens by cre-
ating a uniform layer of paste equal to 0.1 mm. The plates
along with the paste-coated foils were placed inside a
furnace. The structure was heated for a pre-defined time
based on the thickness of the specimen being produced and
then taken out. No post-processing is required to enhance
the mechanical characteristics; therefore, after cooling the
specimen could be used for testing. This setup was
instrumental in determining some key aspects of the pro-
cess and helped in eliminating the guesswork when test
specimens were made using components of the real
machine.
Three tests, including lap-shear test, peel test and
comparative tensile test, were carried out to assess the
capability of CMFM while working with difficult metals.
Specimens were produced using 0.1 mm thick Aluminium
1050 grade foils with a H14 � hard temper bonded by a
special, 80 % zinc and 20 % aluminium by weight, brazing
paste. Tables 1 and 2 show the composition and mechan-
ical properties of the materials used. British and Interna-
tional Standards were followed for the production of parts
and testing. The tests presented in this research analyse the
tensile properties of the specimens and assess the capability
of the process in making parts that provide consistent
results and better material properties compared to con-
ventional subtractive technologies.
2.1 Tensile lap-shear test
Single lap joints using foils of 0.1 mm (t) thickness are not
generally produced which accounts for the lack of specific
standards for such thickness. However, BS EN 1465: 2009
[29] was followed and INSTRON 5582 machine was
operated at a speed of 10 mm/min for testing. The
Table 1 Chemical composition of materials
Materials Chemical composition (in weight %)
Cu Si Fe Mn Mg V O Al
Al 1050 0.05 0.25 0.40 0.05 0.05 0.05 – Balance
Table 2 Mechanical properties of materials
Mechanical properties Materials
Al 1050 H14 Brazing paste
Yield strength (MPa) 105–145 45
Tensile strength (MPa) 120 60
Young’s modulus (GPa) 69 60
Prog Addit Manuf (2016) 1:93–103 95
123
dimensions of the specimen are shown in Fig. 2. The
thickness of the brazing paste was kept at 0.1 mm; there-
fore, the final thickness of the brazed region will be
0.3 mm.
2.2 Peel test
BS EN ISO 11339:2010 [30] was followed for the pro-
duction and testing of the peel specimen. The testing
machine was operated at a speed of 10 mm/min. The
dimensions of the specimen are shown in Fig. 3. The
thickness of the paste was kept at 0.1 mm which makes the
overall thickness of the 150 mm bonded region as 0.3 mm.
2.3 Tensile testing for dog-bone specimens
The dog-bone specimens for tensile test were produced in
accordance with ISO 6892-1 [31]. A specimen was
machined out of a solid aluminium 1050 block and then
compared to parts produced by CMFM. Three specimens
were produced using foils of 0.05, 0.1 and 0.2 mm thick-
ness. The reason for making three different specimens with
foils of different thickness is to analyse their effect. The
dimensions of the specimen are shown in Fig. 4, whereas
Table 3 shows the specification of the three specimens.
For all the specimens the overall thickness was the same
but the individual thickness of the foils differ. The thick-
ness of the paste was again kept at 0.1 mm.
3 Results and discussion
3.1 Results from lap-shear test
The single lap joints produced by CMFM fractured within
the base metal, whereas the brazed zone was unaffected.
This shows that the bond is stronger than the material. Such
a fracture occurred because less force is needed for failure
of the single foil compared to the bonded region that has
twice the thickness. Lap-shear testing is generally affected
by lap joint length, gauge length and asymmetric loading.
These factors were kept in check to ensure consistent
testing conditions. Figure 5 shows the failure modes of the
lap-shear specimens and the test results are shown in
Fig. 6. The failure mode was substrate failure (SF)
according to BS EN ISO 10365:1995 [32].
Two different modes of failure were observed from the
testing:
1. When the fracture was at the base metal adjacent to the
brazed region as seen for S1, S2 and S3. Such a failure
Fig. 2 Dimensions of lap joint [29]
Fig. 3 Dimensions of T-peel specimen [30]
Fig. 4 Dimensions of the dog-bone specimen [31]
Table 3 Specification of the aluminium specimens
Thickness (mm) Number of foils
0.05 18
0.1 14
0.2 9
96 Prog Addit Manuf (2016) 1:93–103
123
is characterized by a smaller displacement range to
failure. Smaller displacement range resulted in smaller
plastic deformation of the foil before failure. Once a
crack was initiated, it ran across the width of the foil to
cause fracture.
2. When a break occurs away from the bonded area.
Stretching of the foil on either ends of the bonded area
led to a fairly larger displacement range in the case of
S4 and S5.
3.1.1 Calculation of lap-shear strength
For industrial applications, tensile lap-shear strength of the
bond or in other words joint tensile strength rð Þ is
important; therefore, lap joints with 10 mm thick Al 1050
plates were made according to BS EN 1465: 2009. The
reason for the mentioned thickness is that anything less
than that resulted in substrate failure, and calculation of
joint tensile strength requires cohesion failure which can be
observed with 10 mm thick plates.
The bond strength (sa) has been calculated as a maxi-
mum shear stress achieved in the bond layer, based on
recorded maximum tensile forces for each tested joint:
sa¼Fmax=Length of the lap joint�width of the substrate:
ð1Þ
The joint tensile strength rð Þ as a maximum tensile
stress transferred crossover the joint:
r¼Fmax=Thickness of the substrate�width of the substrate:
ð2Þ
The values of the bond shear strength and joint tensile
strength are calculated using Eqs. (1) and (2). They are
presented in Table 4.
The experimental calculation yields shear strength of
52.378 MPa (N/mm2) for the bond produced by the brazing
paste used. This value is quite high as compared to some of
the industrial adhesives [33] including a very ductile
polyurethane adhesive (Sikaflex-255 FC with a shear
strength of 8.26 MPa), a very brittle two-component epoxy
adhesive (Araldite� AV138/HV998, with a shear strength
of 30.2 MPa, from Huntsman), and an intermediate two-
component epoxy adhesive (Araldite� 2015, with a shear
strength of 15.9 MPa, from Huntsman). The joint tensile
strength is 62.864 N/mm2 and must be kept in mind while
designing products using CMFM.
3.2 Results from peel test
The bond effectiveness was determined by the peel test. The
specimens had a bond thickness of 0.1 mm; therefore, the
thickness of each specimen becomes 0.3 mm after it has been
produced and is ready for testing. The results of a peel test are
generally influenced by peel angle and peel rate: therefore,
special care was given to keep them constant for the five
specimens. The peel rate was 10 mm/min and the peel angle
was 180�. Figure 7 is the graphical representation and Fig. 8
shows the two failure modes observed during the peel test.
The peel test resulted in two failure modes (Fig. 9):
1. When a specimen shows less and small teeth at the
beginning of the brazed region indicating a strong bond
Fig. 5 Failure modes of lap-shear specimens
0
50
100
150
200
250
300
0 1 2 3 4 5 6 7 8 9
Forc
e (N
)
Displacement (mm)
S1 S2 S3
S4 S5
Fig. 6 Lap-shear test results
Prog Addit Manuf (2016) 1:93–103 97
123
between the two foils, giving a high load ranging from
55 to 50.5 N.
2. When a specimen shows a large number of teeth
resulting in more breaking points that grew as the load
is being applied. Such failures result in comparatively
low force values of around 45.5 N.
Table 5 calculates the average force (from the graph)
and peel strength which is obtained by dividing the maxi-
mum force with the cross-sectional area (0.1 9 25 mm =
2.5 mm2) of each specimen.
The maximum force for the peel tests was much lower
than the maximum force for Al 99.5 which is 250 N
(100 MPa or 100 N/mm2) for a cross-sectional area of
2.5 mm2 (25 9 0.1 mm). These test results differ signifi-
cantly from adhesive bonds that fail uniformly upon the
application of load. When a load was applied to the brazed
specimens, they showed the formation of teeth. The reason
for the formation of teeth is the presence of un-bonded area
indicating an absence of bond between the two foils. The
teeth signify the integrity of the bond produced by CMFM.
Smaller teeth indicate a better bond, whereas larger teeth
signify a poor bond. The presence of un-bonded areas can
be attributed to the application of insufficient force during
brazing, resulting in the paste not being able to penetrate
the tenacious oxide layer at the interface. The peel test
proved useful in understanding the overall bond effec-
tiveness produced by CMFM. The substrate failed by
splitting in layers; therefore, the failure pattern was given
the designation DF (delamination failure).
3.3 Results from dog-bone tensile test
The tensile test was carried out to analyse the behaviour of
the specimens produced by CMFM under tensile forces.
This test clearly demonstrates that bond produced by
CMFM fractures at a higher force value compared to a part
product manufactured from conventional machining
methods. Figure 10 shows the fracture modes of alu-
minium and one of the composite specimens. Figure 11
shows the comparison among the specimens.
Fig. 7 Fracture modes of
aluminium peel test
Table 4 Calculation of joint
tensile strength at 12 mm
overlap length
Specimens Fmax (N) Bond shear strength (MPa) Joint tensile strength (MPa)
S26 15,520 51.73 62.08
S27 15,600 52 62.4
S28 16,000 53.3 64
S29 15,560 51.86 62.24
S30 15,900 53 63.6
Average = 52.378 Average = 62.864
98 Prog Addit Manuf (2016) 1:93–103
123
The maximum force value for the solid aluminium
1050 specimen is 4.483 kN and its curve has well-de-
fined elastic and plastic regions as is expected from an
aluminium alloy. On the other hand, the specimen made
up of 18 layers (0.05 mm thick foils) showed maximum
force value of 4.923 kN, the specimen made up of 14
Fig. 8 Two fracture modes:
a less and regular teeth; b more
and irregular teeth
0
10
20
30
40
50
60
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5
Forc
e (N
)
Displacement (mm)
S1 S2 S3
S4 S5
Fig. 9 Peel test results
Table 5 Peel strength calculation
Specimens Maximum
peeling
force (N)
Average
peeling
force (N)
Peel
strength
(N/mm2)
Type of
failure
S1 45.5 21.73 18.2 Delamination
failureS2 50.5 22.15 20.2
S3 52 23.18 20.8
S4 55 23.11 22
S5 47.5 22 19
Average peel strength = 20.04 N/mm2
Fig. 10 Fracture modes: a Al 1050; b composite specimen
0
1
2
3
4
5
6
0 2 4 6 8 10 12
Forc
e (k
N)
Displacement (mm)
Al 10500.1mm0.2mm0.05mm
Fig. 11 Comparative tensile test
Prog Addit Manuf (2016) 1:93–103 99
123
layers (0.1 mm thick foils) showed value of 4.853 kN
and the specimen made up of nine layers (0.2 mm thick
foils) showed value of 4.754 kN. The first specimen
showed 9.8 %, second showed 8.25 % and the third
showed 6 % higher force values as compared to the
parent aluminium. The values obtained from the tensile
test are plotted in Fig. 12 which shows a linear rela-
tionship between number of layers and force values. It
clearly indicates that the more the number of layers used
for a particular thickness, the more the force required to
fracture the part.
The specimens produced by CMFM showed higher
values because of the formation of an intermetallic
bond between the metal foils and the paste. Aluminium
has a yield strength of 75 MPa and the paste has a
yield strength of 45 MPa. Theoretically, these two
strengths tend to add up with each layer [34]. The same
increase is not obtained through experiments but Fig. 10
shows some addition in strength nonetheless. This
experiment also validates the effect of increased
strength with higher number of layers. 0.2 mm thick
foils were only 9 to produce the 2.7 mm thick speci-
men, giving an increase of 6 %; 0.1 mm thick foils
were 14, giving an increase of 8.25 %; and 0.05 mm
thick foils were 18, thus giving 9.8 % higher values.
This clearly indicates that increase in number of layers
increases the strength of the part produced for the same
cross-sectional area.
Aluminium is a ductile metal; therefore, it shows pro-
nounced elastic and plastic region before fracture. The
composite specimens, on the other hand, show high
strength but a relatively small plastic region due to the
presence of bonds. The intermetallic bonds between the
metal foils and the paste prevent the aluminium layers from
following their ductile nature. The presence of bonds is
also responsible for a much less percentage elongation of
the specimens as compared to the parent aluminium as is
evident from Table 6.
4 Comparison with plate press-soldering process
CMFM is a relatively new process and has shown great
promise in the testing stage. It has huge potential and can
one day be commercialised as a cost-effective metal AM
method capable of producing high-quality metal and
composite parts. This process has a remarkable similarity
with another process which is also in the developmental
stages known as plate press-soldering (PPL). This process
makes use of two massive steel plates as heat source to
provide heat to metal sheets that have been coated with
solder material for joining via contact heating. The process
recommends the utilization of thicker plates to save time
but thicker plates are more expensive than thin metal
sheets. CMFM, on the other hand, can work with metal
sheets of varied thicknesses without any issue as shown in
the results presented in this research. PPL also required a
pre-process for roughening the metal plates by sand
blasting to ensure sufficient adhesion [35], whereas CMFM
can work with metal foils without any surface treatment,
thereby saving time and resources. The heating of the pre-
coated metal sheets is more or less the same in both the
processes. PPL claims to produce parts with lower residual
stress and reduced warping but the data presented are not
substantial enough, whereas we have shown in our previ-
ous work that the thermal stress and strain generated during
the brazing process of CMFM is not high enough to impact
the structural integrity of the final products [36]. PPL also
requires reworking of the resulting part after joining
according to the desired geometry, whereas CMFM does
not require post-processing. The parts produced by CMFM,
after joining, can be used for engineering applications. Two
100 mm long and 7 mm thick spanners, one from alu-
minium and the other a composite of aluminium and cop-
per, produced by the process of CMFM are shown in
Fig. 13 that clearly demonstrates the quality of parts that
CMFM is capable of producing. It also shows that CMFM
is not affected by the staircase effect that a number of metal
foil using methods suffer from [16, 17]. It is because the
layer of paste becomes solid and acts as a stiff sheet
between two metal foils. Furthermore, the parts produced
by CMFM show a much higher tensile strength as
0.2mm, 4.754
0.1mm, 4.853
0.05mm, 4.923
4.74
4.76
4.78
4.8
4.82
4.84
4.86
4.88
4.9
4.92
4.94
0 0.5 1 1.5 2 2.5 3 3.5
Forc
e (k
N)
Samples
Fig. 12 Relationship between strength and foils of different thickness
Table 6 Tensile test values
Specimens Total elongation (%) Ultimate tensile
strength (MPa)
Al 1050 21.5 132.8
0.05 mm 2.6 145.8
0.1 mm 2.8 143.8
0.2 mm 3.1 140.8
100 Prog Addit Manuf (2016) 1:93–103
123
compared to solid parent metals which is something that
the process of PPL is struggling with. The mechanical
testing of the parts produced by PPL show bending tests
performed on them but they are not even comparable to the
base material. A great deal of research is needed to make
this process viable for the production of metal parts with
good mechanical properties. It already has a pre and post
process to get the desired geometry and a secondary pro-
cess to enhance mechanical properties will make the build
of a single part more time-consuming than it already is.
The process of PPL is still in its early stages and requires
more work to ensure reproducible properties of the parts
and choosing a suitable solder material. It is also been
aimed more towards producing large metallic tools rather
than different shapes and geometries. CMFM has shown
reproducible and repeatable results as shown by our pre-
vious work [37] and at this stage is a much more efficient
process compared to PPL.
Scanning electron microscope (SEM) was utilized to
take a closer look at the bonding interface between the
paste and the metal foils (aluminium spanner from
Fig. 13). Figure 14a shows that the layers of metal foil and
paste are diffusing indicating the presence of a larger and
stronger bonded area. A much closer look at the layer of
paste in Fig. 14b showed voids and cavities indicating that
even after good diffusion at the surface of the metal foil,
there are still spaces within the paste layer that could lead
to failure. However, those voids did not affect the
mechanical properties to a large extent as during tensile
testing, the parts made by CMFM proved to be stronger
than solid aluminium. PPL also showed a good diffusion
zone but it was not good enough to enhance the mechanical
properties of parts produced by it.
5 Potential of application
CMFM is a very capable process and can give tough
competition to the current metal AM methods. This process
has the potential to follow the same pattern as three-di-
mensional (3D) printers have gone through in recent years.
3D printers can produce bespoke plastic parts with ease and
CMFM can do the same but with metal parts. There has
always been an emphasis on the production of metal parts
as they provide an environment for testing and show how a
part might behave under different conditions. CMFM can
produce metal parts quickly, cheaply and is much simpler
compared to the rival technologies.
CMFM can work with tough metals like aluminium that
are not easy to join as shown through this research work.
Foils of different thickness are joined using CMFM which
shows the flexibility of the process. It also has the capa-
bility to produce composites in the same way as single
material parts without the use of additional machinery.
This feature makes the process more versatile and
appealing to the metal prototyping sector. CMFM is cap-
able of producing overhanging and complex shaped
geometries with ease without compromising their
mechanical integrity.
The additive manufacturing identified with the genera-
tion of metal parts using the new process can work with an
extensive variety of metals under typical conditions
Fig. 13 Spanners produced by CMFM
Fig. 14 SEM Analysis: a part
at 950; b part at 9200
Prog Addit Manuf (2016) 1:93–103 101
123
regardless of their joining capacities. The feedback that
parts delivered by added substance fabrication techniques
are not sufficiently solid for genuine applications can
without much of a stretch is hushed with the obtained trial
results. Applications can extend from little bespoke parts to
large-scale functional products that can be utilized with no
post handling (Fig. 13). They include but are not limited to
mechanical workshop tools, motor industry (engines and
their parts), large moulds, etc. This process is an addition to
the list of metal AM methods and has distinct advantages
including higher lap shear strength, peel strength, corrosion
resistance [37], tensile strength, etc. and is well-suited for
single builds.
In terms of production time, CMFM can produce parts at a
much faster rate. Figure 13 shows two dog-bone specimens
made by CMFM and DMLS machine (EOS M290). A visual
inspection shows that the part produced by the former pro-
cess (Fig. 15a) has a better surface finish and as it has already
been established, possesses a higher lap shear and tensile
strength. The part made by the latter process has a grainy
finish (Fig. 15b) and at some areas shows impressions of the
build material structure (Fig. 15c). The dog-bone specimen
with a thickness of 2.7 mm using foils of 0.1 mm thickness
can be made in approximately 30 min. The same part in
DMLS takes a longer time. Pre-processing of the part
requires 45 min for the gas to fill the build chamber and 20
min for setting up and levelling of the build platform. The
actual part build takes 40 min including a 4 mm thick support
structure; otherwise the actual part would not adhere to the
surface. Post-processing includes 15 min for powder
removal after the build, 10 min for removing the part from
the build platform and the support structure from the part. It
clearly shows the inefficiency of a well-established metal
AM method when it comes to performing single builds.
The parts made by DMLS are not without inaccuracies.
The accuracy depends on a number of factors including
powder material, grain size, layer thickness, etc. and is an
issue in all three axes. The achievable part accuracy
depends on which powder material is used, varying from
about ±50 lm for Direct-Metal 20 to about ±100 lm for
Direct-Steel 50. The resolution in the vertical direction
(perpendicular to the layers) is determined by the layer
thickness. For Direct-Metal 20, Direct-Steel 20 and
Direct-Steel H20 it is typically 0.02 mm, and for Direct-
Metal 50 and Direct-Steel 50 it is 0.05 mm. For Direct-
Metal 50 (grain size approx. 50 mm) accuracies of
±(0.05 % ? 50 mm) are obtainable. Due to the slightly
higher shrinkage during exposure and the smaller powder
particle size the obtainable accuracy for Direct-Metal 20,
Direct-Steel 20, Direct-Steel H20 and Direct-Steel 50 is
±(0.07 ? 50 lm) [38]. In the case of CMFM, there is not a
considerable amount of inaccuracy/deformation in the
X and Y axes [36]. The accuracy of the Z-axis depends on
the uniformity/non-uniformity of the metal foil used and
varies according to that factor. The accuracy in the vertical
direction (Z axis) depends on the layer thickness and is
found to be ±100 micron.
6 Conclusions
The results presented in this paper show the effectiveness of
CMFM. The lap-shear testing showed two modes of failure
but the failure mode was always substrate failure whereas the
bonded area remained unaffected and intact. The peel test
also exhibited two modes of failure with the presence of
shorter and longer teeth signifying the presence of an
effective and un-effective bond, respectively. The strength of
the dog-bone specimens produced by CMFM is higher than
the specimen produced by conventional machining method.
It was argued and validated that the strength increases with
increase in the number of layers for a particular cross-sec-
tional area. The specimen produced by 0.05 mm thick layers
showed 9.8 %, 0.1 mm thick layers showed 8.25 % and the
specimen produced by 0.2 mm thick layers showed 6 %
higher fracture values than the parent aluminium. This is an
added feature that the process can offer for the production of
metal parts using foils of different thickness.
Acknowledgments The authors would like to thank Anglia Ruskin
University for providing the equipment and testing facilities for car-
rying out the research.
References
1. Dickens PM (1995) Research developments in rapid prototyping.
Proc Inst Mech Eng Part B J Eng Manuf 209(4):261–266Fig. 15 Parts made by different processes: a final part by CMFM;
b final part by DMLS; c Part by DMLS with support structure
102 Prog Addit Manuf (2016) 1:93–103
123
2. Himmer T, Nakagawa T, Noguchi H (1997). Stereolithography of
ceramics. In: international solid freeform fabrication symposium,
Austin, pp 363–369
3. Kai CC (1994) Three-dimensional rapid prototyping technologies
and key development areas. Comput Control Eng J 5(4):200–206
4. Pham DT, Gault RS (1998) A comparison of rapid prototyping
technologies. Int J Mach Tools Manuf 38(10):1257–1287
5. Zhang Y, He X, Du S, Zhang J (2001) Al2O3 ceramics prepa-
ration by LOM (laminated object manufacturing). Int J Adv
Manuf Technol 17(7):531–534
6. Calignano F, Manfredi D, Ambrosio EP, Iuliano L, Fino P (2013)
Influence of process parameters on surface roughness of alu-
minum parts produced by DMLS. Int J Adv Manuf Technol
67(9–12):2743–2751
7. Cormier D, Harrysson O, West H (2004) Characterization of H13
steel produced via electron beam melting. Rapid Prototyp J
10(1):35–41
8. Mitchell A (1999) The electron beam melting and refining of
titanium alloys. Mater Sci Eng A 263(2):217–223
9. Simchi A (2006) Direct laser sintering of metal powders:
mechanism, kinetics and microstructural features. Mater Sci Eng
A 428(1):148–158
10. Simchi A, Petzoldt F, Pohl H (2003) On the development of
direct metal laser sintering for rapid tooling. J Mater Process
Technol 141(3):319–328
11. Sonmez FO, Hahn HT (1998) Thermomechanical analysis of the
laminated object manufacturing (LOM) process. Rapid Prototyp J
4(1):26–36
12. Ponader S, Vairaktaris E, Heinl P, Wilmowsky CV, Rottmair A,
Korner C, Singer RF, Holst S, Schlegel KA, Neukam FW,
Nkenke E (2008) Effects of topographical surface modifications
of electron beam melted Ti-6Al-4 V titanium on human fetal
osteoblasts. J Biomed Mater Res Part A 84(4):1111–1119
13. Murr LE, Gaytan SM, Ceylan A, Martinez E, Martinez JL,
Hernandez DH, Machado BI, Ramirez DA, Medina F, Collins S,
Wicker RB (2010) Characterization of titanium aluminide alloy
components fabricated by additive manufacturing using electron
beam melting. Acta Mater 58(5):1887–1894
14. Khaing MW, Fuh JYH, Lu L (2001) Direct metal laser sintering
for rapid tooling: processing and characterisation of EOS parts.
J Mater Process Technol 113(1):269–272
15. Mueller B, Kochan D (1999) Laminated object manufacturing for
rapid tooling and patternmaking in foundry industry. Comput Ind
39(1):47–53
16. Prechtl M, Otto A, Geiger M (2005) Rapid tooling by laminated
object manufacturing of metal foil. Trans Tech 6–8:302–312
17. Prechtl M, Pursche L, Otto A (2004) System technology and data
preparation for automated laser assisted stacking of metal foil. In:
Proceedings of the 4th international conference on laser assisted
net shape engineering—LANE, Germany, pp. 601–610
18. Armstrong KB (1997) Long-term durability in water of alu-
minium alloy adhesive joints bonded with epoxy adhesives. Int J
Adhes Adhes 17(2):89–105
19. Kinloch AJ (2003) Toughening epoxy adhesives to meet today’s
challenges. MRS Bull 28(06):445–448
20. Kong CY, Soar RC (2005) Fabrication of metal–matrix com-
posites and adaptive composites using ultrasonic consolidation
process. Mater Sci Eng A 412(1):12–18
21. Kong CY, Soar RC, Dickens PM (2002) An investigation of the
control parameters for aluminum 3003 under ultrasonic consoli-
dation. In: proceedings of the 13th solid freeform fabrication
symposium, pp. 199–210
22. Kong CY, Soar RC, Dickens PM (2003) Characterisation of
aluminium alloy 6061 for the ultrasonic consolidation process.
Mater Sci Eng A 363(1):99–106
23. Facchini L, Magalini E, Robotti P, Molinari A, Hoges S, Wis-
senbach K (2010) Ductility of a Ti-6Al-4 V alloy produced by
selective laser melting of prealloyed powders. Rapid Prototyp J
16(6):450–459
24. Gu D, Shen Y (2006) WC–Co particulate reinforcing Cu matrix
composites produced by direct laser sintering. Mater Lett
60(29):3664–3668
25. Kumar S, Kruth JP (2010) Composites by rapid prototyping
technology. Mater Des 31(2):850–856
26. Butt J, Mebrahtu H, Shirvani H (2015) Rapid prototyping by heat
diffusion of metal foil and related mechanical testing. Int J Adv
Manuf Tech. doi:10.1007/s00170-015-7882-8
27. Butt J, Mebrahtu H, Shirvani H (2014) A novel rapid prototyping
process for the production of metal parts, ‘‘proceedings of the
second international conference on advances in civil, structural
and mechanical engineering, Birmingham, pp 26–29. doi:
10.15224/978-1-63248-054-5-45
28. ASM Metals Handbook (1985) Heat Treating, vol 4, 9th edn.
American Society for Metals, Metals Park, pp 763–774
29. BS EN 1465: 2009 Adhesives—determination of tensile lap-shear
strength of bonded assemblies
30. BS EN ISO 11339:2010 Adhesives—T-peel test for flexible-to-
flexible bonded assemblies
31. BS EN ISO 6892-1:2009 Metallic materials—tensile testing, part
1: method of test at ambient temperature
32. BS EN ISO 10365:1995 Adhesives—designation of main failure
patterns
33. Da Silva LF, Carbas RJC, Critchlow GW, Figueiredo MAV,
Brown K (2009) Effect of material, geometry, surface treatment
and environment on the shear strength of single lap joints. Int J
Adhes Adhes 29(6):621–632
34. Hassan Shirvani (2000) Honeycomb structure, method of forming
a honeycomb structure and method of joining aluminium or
aluminium alloy bodies. Anglia Polytechnic University, GB
Patent: WO0033997 (A1)
35. Strasser C, Prihodovsky A, Ploshikhin V (2015). Plate press-sol-
dering for the manufacture of large metallic tools. https://www.
rtejournal.de/ausgabe12/4242/view?set_language=en. Accessed
29 Feb 2016
36. Butt J, Mebrahtu H, Shirvani H (2015) Thermo-mechanical
analysis of dissimilar al/cu foil single lap joints made by com-
posite metal foil manufacturing. World Acad Sci Eng Technol Int
J Mech Aerospa Ind Mech Manuf Eng 10(1):41–46
37. Butt J, Mebrahtu H, Shirvani H (2015) Peel and tensile test
investigation of aluminium 1050 foil parts made with a new
additive manufacturing process. Int J Rapid Manuf 5(1):95–115
38. Syvanen T (1996) Direct sintering of metals material and process
development M.Sc. Thesis, Tampere University of Technology
Prog Addit Manuf (2016) 1:93–103 103
123
